Third Generation Photovoltaics

Third Generation Photovoltaics

(Parte 1 de 6)

springer series in photonics 12 springer series in photonics 12 springer series in photonics

Series Editors: T. Kamiya B. Monemar H. Venghaus Y. Yamamoto

The Springer Series in Photonics covers the entire field of photonics, including theory, experiment, and the technology of photonic devices. The books published in this series give a careful survey of the state-of-the-art in photonic science and technology for all the relevant classes of active and passive photonic components and materials. This series will appeal to researchers, engineers, and advanced students.

1 Advanced Optoelectronic Devices By D. Dragoman and M. Dragoman

2 Femtosecond Technology

Editors: T. Kamiya, F. Saito, O. Wada, andH .Yajima

3 Integrated Silicon Optoelectronics By H. Zimmermann

4 Fibre Optic Communication Devices Editors: N. Grote and H. Venghaus

5 Nonclassical Light from Semiconductor Lasers andLEDs By J. Kim, S. Somani, and Y. Yamamoto

6 Vertical-Cavity Surface-Emitting

Laser Devices By H. Li and K. Iga

7 Active Glass for Photonic Devices

Photoinduced Structures and Their Application Editors: K. Hirao, T. Mitsuyu, J. Si, andJ .Q iu

8 Nonlinear Photonics

Nonlinearities in Optics, Optoelectronics and Fiber Communications By Y. Guo, C.K. Kao, E.H. Li, and K.S. Chiang

9 Optical Solitons in Fibers

Third Edition By A. Hasegawa and M. Matsumoto

10 Nonlinear Photonic Crystals Editors: R.E. Slusher and B.J. Eggleton

1 Waveguide Nonlinear-Optic Devices By T. Suhara and M. Fujimura

12 Third Generation Photovoltaics

Advanced Solar Energy Conversion By M.A. Green

13 Thin Film Solar Cells

Next Generation Photovoltaics and Its Application Editor: Y. Hamakawa

M.A. Green

Third Generation Photovoltaics

Advanced Solar Energy Conversion

With 63 Figures

ProfessorM artinA .G reen

University of South Wales Centre of Excellence for Advanced Silicon Photovoltaics and Photonomics Sydney, NSW, 2052, Australia

Series Editors:

Professor Takeshi Kamiya

Ministry of Education, Culture, Sports, Science and Technology, National Institution for Academic Degrees, 3-29-1 Otsuka, Bunkyo-ku, Tokyo 112-0012, Japan

ProfessorB oM onemar

Department of Physics and Measurement Technology Materials Science Division Linkoping University 58183 Linkoping, Sweden

Dr. Herbert Venghaus

Heinrich-Hertz-Institut fur Nachrichtentechnik Berlin GmbH Einsteinufer 37 10587 Berlin, Germany

Professor Yoshihisa Yamamoto

Stanford University Edward L. Ginzton Laboratory Stanford, CA 94305, USA

ISBN-10 3-540-26562-7 Springer Berlin Heidelberg New York ISBN-13 978-3-540-26562-7 Springer Berlin Heidelberg New York

ISBN 3-540-40137-7 Springer Berlin Heidelberg New York ( -Cover Edition)

Libraryo fC ongressC ontrolN umber:

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproductiono nm icrofilmo ri na nyo therw ay,a nds toragei nd atab anks.D uplicationo ft hisp ublicationo r parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable to prosecution under the German Copyright Law.

Springer is a part of Springer Science+Business Media. springer.com

© Springer-Verlag Berlin Heidelberg 2003, 2006 Printed in The Netherlands

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Data conversion: Marianne Schillinger-Dietrich, Berlin

Cover production: design & production GmbH, Heidelberg Printed on acid-free paper SPIN: 11499992 543210

57/3100/SPI Publisher Services

Cover concept: eStudio Calamar Steinen roductionSPI Publisher ServicesP:

To Judy, Brie and Morgan To Judy, Brie and Morgan

Preface

This text has its origins in my personal perceptions of how the photovoltaic industry is likely to develop as it expands to its full potential.

After my group’s work in the early 1980s on improving silicon laboratory solar cell performance, we were able to capture most of this improvement in a commercially viable sequence, via the buried contact, laser grooved cell. Through the efforts of BP Solar, this has become one of the most successfully commercialised new cell technologies since then, with sales certain to exceed US$1 billion by 2010. This technology addresses the high material cost of “firstgeneration”, silicon wafer-based photovoltaics by improving power out per unit investment in such material.

My group then was given the opportunity in the late 1980s to broaden its program into “second-generation” thin-film approaches, more directly addressing the issue of material costs. Although the then-favoured thin-film options, amorphous silicon, copper indium diselenide and cadmium telluride, all had their strengths, I believed there were quite fundamental limitations with each due to stability, resource availability and/or toxicity. Our previous success with silicon gave us confidence that we could develop a more desirable thin-film polycrystalline-silicon-on-glass technology almost from “scratch”. A decade later, with rapidly increasing pilot-line module efficiencies being demonstrated by Pacific Solar with this “silicon on glass” approach, it became clear we had met our aim of developing a more viable thin-film option. This “second-generation” technology is capable of supporting the growth of the photovoltaic industry to beyond 2020, due to the quantum cost reduction it offers by eliminating wafers.

We then began to think about how this new technology might develop with time. Incremental refinements in material quality and device design were likely to increase efficiencies to close to 15%, comparable to the best presently with “firstgeneration” modules. We realised that, post-2020, with photovoltaics a large, profitable industry, there would be pressure to increase performance beyond this, since “second-generation” technology by then would be constrained by its own material costs. Just as the microelectronics industry relentlessly pushes towards smaller feature size to reduce costs, a mature photovoltaics industry would push towards ever-increasing conversion efficiency!

Tandem cells, where cells of different bandgap material are stacked on top of one another, offer a well-proved approach to increased efficiency. However, tandems involving compound semiconductors on top of thin-film silicon would not make a great deal of sense. There would be no compelling reason for using silicon in such a device, but rather compound material similar to that in the overlying device. Each cell added to a tandem stack also increases processing complexity and sensitivity to changes in the spectral content of sunlight. Were

VIII Preface there alternative, more elegant approaches to increased performance, perhaps more compatible with the thin silicon on glass technology we had helped develop?

Both to answer this question and to meet the more mundane need to differentiate future research from work funded in the past, we were led to the concept of a third generation of photovoltaics. This would be differentiated from the two earlier generations by higher performance potential than from single junction devices. Other key criteria were that it use thin-films, for low material costs, and abundant, non-toxic materials. Although silicon is ideal in this regard, progress with molecularly based systems such as organic and dye sensitised cells and with nanostructural engineering in general, suggested other comparably attractive material systems may become available by 2020.

The first phase in our attempts to identify third generation candidates was to gain a clear understanding of the strengths and weaknesses of approaches suggested in the past for improving performance. It was also hoped that this reexamination might stimulate new ideas. This book documents the results of this phase. Taking a very broad view of photovoltaics, almost as broad as “electricity from sunlight”, advanced photovoltaic options are analysed self-consistently with key features and challenges for successful implementation assessed. Although radiative inefficiencies readily can be incorporated, the main focus is on performance in the radiative limit. The rationale for this is that all successful photovoltaic devices must evolve towards this limit as argued above.

I would like to thank all who have stimulated my interest in photovoltaics since the early days, either by direct contact or by published work. I particularly thank those who took my postgraduate course on advanced photovoltaics during 2000, acting as guinea pigs for developing the text’s first draft. Andrew Brown, Nils Harder and Holger Neuhaus deserve special mention for constantly challenging the material presented and for several graphs and tables in the text. I also thank Richard Corkish, Thorsten Trupke and Stuart Wenham and the high profile researchers on the Advisory Committee of the Centre formed to explore third generation options, particularly the longest serving members, Professors Antonio Luque, Hans Queisser and Peter Würfel. As the reader will note, the book also benefits from their past work. I also thank the Humboldt Foundation for a Senior Research Award and Professors Ernst Bücher, Ulrich Gösele and Rudolph Hezel for hosting associated visits during 2001 and 2002 where, amongst other activities, the manuscript was finalised. Finally, I thank Jenny Hansen for tireless efforts in producing diagrams plus many drafts of the text and Judy Green for support and companionship over the period this book was developed.

Bronte, SydneyMartin A. Green January, 2003

Table of Contents

1.1 “Twenty-Twenty Vision”1
1.2 The Three Generations1
1.3 Outline of Options4

1 Introduction

2.1 Introduction7
2.2 Black-Body Radiation7
2.3 Black-Body in a Cavity10
2.4Angular Dependence of Emitted Radiation1
2.5 Direct and Diffuse Efficiencies15
2.6 Black-Body Emission Properties16

2Black-Bodies, White Suns

3.1 Introduction21
3.2 Energy and Entropy Conservation21
3.3 Carnot Efficiency2
3.4 Landsberg Limit24
3.5 Black-Body Limit25
3.6 Multi-Colour Limit27
3.7 Non-Reciprocal Systems39
3.8 Ultimate System30
3.9 Omnidirectional Global Converters31
3.10 Summary32

3Energy, Entropy and Efficiency

4.1 Efficiency Losses35
4.2 Shockley-Queisser Formulation38
4.3Hot Photons (Chemical Potential of Light)40
4.4 Einstein Coefficients43
4.5 Photon Boltzmann Equation45
4.6 General Cell Analysis49
4.7 Lasing Conditions50
4.8 Photon Spatial Distributions51
4.9 Effect of Sample Thickness54

4Single Junction Cells 4.10Thermodynamics of Single Junction Cell..........................5

X Table of Contents

5.1 Spectrum Splitting and Stacking59
5.2 Split-Spectrum Cells60
5.3 Stacked Cells61
5.4 Two Terminal Operation63
5.5 Infinite Number of Cells64
5.6 Approximate Solution65
5.7Thermodynamics of the Infinite Stack6

5Tandem Cells

6.1 Introduction69
6.2 Relevant Time Constants69
6.3 Ross and Nozik’s Analysis72
6.4Simplification for EG = 07
6.5 Würfel’s Analysis7
6.6 Possible Low Dimensional Implementation79
7.1 Introduction81
7.2 Multiple-Carrier Photon Emission82
7.3 Limiting Performance85
7.4 Comparison with Würfel’s Analysis85
7.5 Possible Implementation86
7.6 Generalised Analysis86
7.7 Raman Luminescence8

7Multiple Electron-Hole Pairs per Photon

8.1 Introduction95
8.2 3-Band Cell97
8.3 Photon Absorption Selectivity98
8.3.1 Finite Bandwidths98
8.3.2 Graded Absorption Coefficients100
8.3.3 Spatial Absorption Partitioning100
8.4 Absorption Leakage Loss102
8.5 Other Possible Multigap Schemes104
8.6 Impurity Photovoltaic Effect106
8.7 Up- and Down-Conversion107

8Impurity Photovoltaic and Multiband Cells

9.1 Introduction1
9.2 Solar Thermal Conversion113
9.3 Thermophotovoltaic Conversion114
9.3.1 Black-Body Source114
9.3.2 With Narrow Passband Filter116
9.3.3 Solar Conversion: Cell/Receiver117

9Thermophotovoltac and Thermophotonic Conversion 9.4 Thermophotonics ............................................... 118

9.4.1 Case with Filters118
9.4.2 Without Filter122
10 Conclusions125

Table of Contents XI

A: Greek Alphabet127
B: Physical Constants129
C: Fermi-Dirac and Bose-Einstein Integrals131
D: List of Symbols137
E: Quasi-Fermi Levels139
F: Solutions to Selected Problems147

1 Introduction

1.1 “Twenty-Twenty Vision”

Since the early days of terrestrial photovoltaics, many have thought that “first generation” silicon wafer-based solar cells eventually would be replaced by a “second generation” of inherently much less material intensive thin-film technology, probably also involving a different semiconductor. Historically, cadmium sulphide, amorphous silicon, copper indium diselenide, cadmium telluride and now thin-film silicon have been regarded as key thin-film candidates. Since any mature solar cell technology must evolve to the stage where cost is dominated by that of the constituent material, be it silicon wafers or glass sheet, it seems that high power output per unit area is the key to the lowest possible future manufacturing costs. Such an analysis makes it likely that photovoltaics, in its most mature form, will evolve to a “third generation” of high-efficiency, thin-film technology. By high-efficiency, what is meant is energy conversion values double or triple the 15-20% range presently targeted, closer to the thermodynamic limit upon solar conversion of 93%.

Tandem or stacked cells provide the best known example of how such high efficiency might be achieved. In this case, conversion efficiency can be increased merely by adding more cells of different bandgap to a stack, at the expense of increased complexity and spectral sensitivity. However, as opposed to this “serial” approach, better-integrated “parallel” approaches are possible that offer similar efficiency to even a stack involving an infinite number of such tandem cells. These alternatives will become increasingly feasible with the likely evolution of materials technology over the decades to 2020. This book discusses a range of these options systematically as well as paths to practical implementation. By clearly defining these options and identifying their strengths, weaknesses and areas where further work is required, their development may be accelerated.

1.2 The Three Generations

Most solar cells sold in 2003 were based on silicon wafers, so-called “first generation” technology (Fig. 1.1). As this technology has matured, its economics have become dominated increasingly by the costs of starting materials already

2 1 Introduction

Fig. 1.1: Example of “first-generation” wafer-based technology (BP Solar Saturn Module using UNSW buried contact technology).

made in high volume and hence with little potential for cost reduction, such as silicon wafers, toughened low-iron glass cover sheet and other encapsulants. From experience with other technologies, this trend is expected to continue as the photovoltaic industry continues to mature. For example, a recent study (Bruton et al. 1997; Bruton 2002) of costs of manufacturing in a 500 MW/year production facility, about 10 times larger than the largest facilities in 2003, suggests material costs would account for over 70% of total manufacturing costs. The study therefore predicts lowest cost for high-efficiency processing sequences, provided these do not unduly complicate cell processing . Nonetheless, module efficiency above 16% is not contemplated even in this futuristic scenario.

For a prolonged period extending from the early 1980s, it has seemed that the photovoltaic industry has been on the verge of switching to a “second generation” of thin-film solar cell technology. Regardless of the semiconductor involved, thinfilm technology offers prospects for a large reduction in material costs by eliminating the costs of the silicon wafer. Thin-film technology also offers other advantages, such as the increased size of the unit of manufacturing. This increases from the area of a silicon wafer (~ 100 cm2) to that of a glass sheet (~ 1 m2), about 100 times larger. On the efficiency front, with time, most would expect this “second generation” technology steadily to close the gap between its performance and that of “first generation” product.

As thin-film “second generation” technology matures, costs again will become progressively dominated by those of the constituent materials, in this case, the top

1.2 The Three Generations 3 cover sheet and other encapsulants (Woodcock et al. 1997). There will be a lower limit on such costs that, when combined with likely attainable cell efficiency (15% or 150 peak watts/m2), determines the lower limit on photovoltaic module and, hence, electricity generation costs.

One approach to progress further is to increase conversion efficiency substantially. In principle, sunlight can be converted to electricity at an efficiency close to the Carnot limit of 95% for the sun modelled as a black-body at 6000 K and a 300 K cell (Chap. 2). This is in contrast to the upper limit, 31% on the same basis, upon the conversion efficiency of a single junction solar cell, as would limit silicon wafer and most present thin-film devices. This suggests the performance of solar cells could be improved 2-3 times if fundamentally different underlying concepts were used in their design, ultimately to produce a “third generation” of high performance, low-cost photovoltaic product.

(Parte 1 de 6)

Comentários